10
496 http://journals.tubitak.gov.tr/botany/ Turkish Journal of Botany Turk J Bot (2016) 40: 496-505 © TÜBİTAK doi:10.3906/bot-1511-22 Ultrastructural observations in somatic embryogenesis of natural tetraploid Trifolium pratense L. Hatice ÇÖLGEÇEN 1, *, Hatice Nurhan BÜYÜKKARTAL 2 , Ufuk KOCA-ÇALIŞKAN 3 1 Department of Biology, Faculty of Arts and Science, Bülent Ecevit University, Zonguldak, Turkey 2 Department of Biology, Faculty of Science, Ankara University, Ankara, Turkey 3 Department of Pharmacognosy, Faculty of Pharmacy, Gazi University, Ankara, Turkey * Correspondence: [email protected] 1. Introduction Legumes are one of the most important food crops due to their high nutritional value and medicinal properties, as well as their use in livestock feeding. Trifolium pratense L. (red clover), one of the most studied diploid species, is widely grown worldwide. In addition to the classical methods, in vitro methods have a significant place in agricultural studies. Genetic selection can be made by production of somatic embryos. Previously, a 40% somatic embryo production rate was achieved from cell suspension cultures of T. pratense Arlington using a basal medium supplemented with 0.01 mg/L 2,4-D and 2 mg/L adenine (Phillips and Collins, 1980). Direct somatic embryogenesis on immature zygotic embryos in vitro has been confirmed for T. pratense and T. repens. For both species, direct embryo cloning was achieved on an appropriate basal medium supplemented with 0.05 mg/L BAP and 1 g/L yeast (Maheswaran and Williams, 1984, 1985, 1986). Direct somatic embryogenesis was reported in mesophyll protoplast cultures of T. pratense using the nutrient medium of Kao and Michayluk supplemented with 0.5 mg/L 2,4-D, 0.5 mg/L kinetin, and 0.1% casein hydrolysate (Radionenko et al., 1994). Available studies in the literature are limited to diploid varieties. Natural tetraploid T. pratense L. is an important plant that is grown in Turkey. e plant has high biomass efficiency; however, seed settings and hard seed problems hinder production of T. pratense. Bakar Büyükkartal (2008) observed that a mature embryo sac was formed in 18% of ovules examined, but the seed setting ratio was remarkably low (5.8%) in natural tetraploid T. pratense. Previous plant regeneration of the natural tetraploid T. pratense could have been realized only through the apical meristem calli (84.56%) (Çölgeçen and Toker, 2008). In order to proceed to the stage of production, other regeneration methods need to be explored. An effective and efficient in vitro production method is needed for utilizing natural tetraploid T. pratense in different studies (for example, phytoestrogens production in bioreactor, genetic selection). One method for in vitro Abstract: Previous reports of plant regeneration of natural tetraploid T. pratense L. ‘Elçi’ could be realized only through the apical meristem calli. In order to proceed to the production stage, other regeneration methods need to be tried. Aseptic seedlings were used for the production of somatic embryos through various 2,4-D and kinetin trials. Nonuniform external callus cells with translucent cytoplasm were observed in various developmental stages of somatic embryos. Beneath these cells, there were uniformly aligned, dark- stained embryo cells with dense cytoplasm. Despite the similar developmental stages and cell characteristics of zygotic and somatic embryos, the walls of somatic embryo cells revealed a highly wavy pattern. e nucleus generally contained only one nucleolus, which was spherical, dark stained, and electron-dense. Electron-dense droplets were seen in vacuoles. e cytoplasm consisted of starch- containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, dictyosomes, lipid, and protein bodies. In some of the somatic embryos at the globular and heart stages, vacuole or electron-translucent zones were observed in the nucleolus. Additionally, a few embryo degenerations were recorded during developmental stages of the zygotic embryo. For the first time, the somatic embryos of natural tetraploid T. pratense were produced from hypocotyl (85%), cotyledon (75%), and apical meristem (60%) explants in 0.3 mg/L 2,4-D and 2 mg/L kinetin-containing MS medium. Our study developed an effective and efficient in vitro production method for using natural tetraploid T. pratense in biotechnological studies. Key words: Trifolium, red clover, callus, somatic embryo, transmission electron microscope (TEM) Received: 20.11.2015 Accepted/Published Online: 25.03.2016 Final Version: 19.07.2016 Research Article

Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

496

http://journals.tubitak.gov.tr/botany/

Turkish Journal of Botany Turk J Bot(2016) 40: 496-505© TÜBİTAKdoi:10.3906/bot-1511-22

Ultrastructural observations in somatic embryogenesis ofnatural tetraploid Trifolium pratense L.

Hatice ÇÖLGEÇEN1,*, Hatice Nurhan BÜYÜKKARTAL2, Ufuk KOCA-ÇALIŞKAN3

1Department of Biology, Faculty of Arts and Science, Bülent Ecevit University, Zonguldak, Turkey 2Department of Biology, Faculty of Science, Ankara University, Ankara, Turkey

3Department of Pharmacognosy, Faculty of Pharmacy, Gazi University, Ankara, Turkey

* Correspondence: [email protected]

1. IntroductionLegumes are one of the most important food crops due to their high nutritional value and medicinal properties, as well as their use in livestock feeding. Trifolium pratense L. (red clover), one of the most studied diploid species, is widely grown worldwide. In addition to the classical methods, in vitro methods have a significant place in agricultural studies. Genetic selection can be made by production of somatic embryos. Previously, a 40% somatic embryo production rate was achieved from cell suspension cultures of T. pratense Arlington using a basal medium supplemented with 0.01 mg/L 2,4-D and 2 mg/L adenine (Phillips and Collins, 1980). Direct somatic embryogenesis on immature zygotic embryos in vitro has been confirmed for T. pratense and T. repens. For both species, direct embryo cloning was achieved on an appropriate basal medium supplemented with 0.05 mg/L BAP and 1 g/L yeast (Maheswaran and Williams, 1984, 1985, 1986). Direct somatic embryogenesis was reported in mesophyll protoplast cultures of T. pratense using the

nutrient medium of Kao and Michayluk supplemented with 0.5 mg/L 2,4-D, 0.5 mg/L kinetin, and 0.1% casein hydrolysate (Radionenko et al., 1994). Available studies in the literature are limited to diploid varieties.

Natural tetraploid T. pratense L. is an important plant that is grown in Turkey. The plant has high biomass efficiency; however, seed settings and hard seed problems hinder production of T. pratense. Bakar Büyükkartal (2008) observed that a mature embryo sac was formed in 18% of ovules examined, but the seed setting ratio was remarkably low (5.8%) in natural tetraploid T. pratense.

Previous plant regeneration of the natural tetraploid T. pratense could have been realized only through the apical meristem calli (84.56%) (Çölgeçen and Toker, 2008). In order to proceed to the stage of production, other regeneration methods need to be explored. An effective and efficient in vitro production method is needed for utilizing natural tetraploid T. pratense in different studies (for example, phytoestrogens production in bioreactor, genetic selection). One method for in vitro

Abstract: Previous reports of plant regeneration of natural tetraploid T. pratense L. ‘Elçi’ could be realized only through the apical meristem calli. In order to proceed to the production stage, other regeneration methods need to be tried. Aseptic seedlings were used for the production of somatic embryos through various 2,4-D and kinetin trials. Nonuniform external callus cells with translucent cytoplasm were observed in various developmental stages of somatic embryos. Beneath these cells, there were uniformly aligned, dark-stained embryo cells with dense cytoplasm. Despite the similar developmental stages and cell characteristics of zygotic and somatic embryos, the walls of somatic embryo cells revealed a highly wavy pattern. The nucleus generally contained only one nucleolus, which was spherical, dark stained, and electron-dense. Electron-dense droplets were seen in vacuoles. The cytoplasm consisted of starch-containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, dictyosomes, lipid, and protein bodies. In some of the somatic embryos at the globular and heart stages, vacuole or electron-translucent zones were observed in the nucleolus. Additionally, a few embryo degenerations were recorded during developmental stages of the zygotic embryo. For the first time, the somatic embryos of natural tetraploid T. pratense were produced from hypocotyl (85%), cotyledon (75%), and apical meristem (60%) explants in 0.3 mg/L 2,4-D and 2 mg/L kinetin-containing MS medium. Our study developed an effective and efficient in vitro production method for using natural tetraploid T. pratense in biotechnological studies.

Key words: Trifolium, red clover, callus, somatic embryo, transmission electron microscope (TEM)

Received: 20.11.2015 Accepted/Published Online: 25.03.2016 Final Version: 19.07.2016

Research Article

Page 2: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

497

plant regeneration is somatic embryogenesis. Although, there is much research on somatic embryogenesis in Trifolium species and Fabaceae species like soy bean and alfalfa, no study has been completed for natural tetraploid T. pratense. For instance, somatic embryos were produced from leaf-derived protoplasts of Medicago sativa (Monteiro et al., 2003). A histological analysis was conducted on the somatic embryos obtained from cell-suspension cultures of Vigna unguiculata (Ramakrishnan et al., 2005). The embryological studies in natural tetraploid T. pratense examined the ultrastructure of the zygotic embryo and endosperm development, ultrastructure of the seed coat throughout developmental stages, apoximis, and factors causing poor seed setting (Algan and Bakar, 1996, 1997, 1999a, 1999b; Bakar and Algan, 1998).

The aim of this study was to develop an effective method for producing somatic embryos from aseptic seedlings by using different plant growth regulators and to examine the ultrastructural properties of somatic embryos of natural tetraploid T. pratense.

2. Materials and methods2.1. Plant material and culture conditionsThis study examined natural tetraploid E2 type (2n = 4x = 28 chromosomes) Trifolium pratense L. collected from Tortum in the vicinity of Erzurum, Turkey by Elçi (1982). The E2-type natural tetraploid T. pratense L. was grown in the experimentation gardens of Ankara University’s Department of Biology in the Faculty of Science.

Fifteen-day-old aseptic seedlings with a unifoliate primary leaf were used as the explant source. Seeds were first sterilized in 96% ethanol for 1 min and then transferred to 10% sodium hypochlorite solution for 10 min (commercial sodium hypochlorite was used in the sterilization process). Then seeds were rinsed 3 times in autoclaved, distilled water. After being scarified with autoclaved sandpaper, seeds were germinated on hormone-free MS medium (Murashige and Skoog, 1962). All the samples were incubated at 22–24 °C with a 16-/8-h photoperiod (irradiance of 42 µmol m–2 s–1 provided by cool-white fluorescent tubes). Then 25 explants were taken per hypocotyl (0.5–1 cm), cotyledon (whole and in two fragments), apical meristem (1 mm), epicotyl (0.5–1 cm),

and young primary leaves (whole and divided into two fragments) of aseptically grown seedlings. The explants were cultured in petri dishes in the dark (100 mm × 15 mm) (Çölgeçen and Toker, 2008). The embryogenic calli were subcultured onto the same media after 8 weeks.

Embryoid callus was produced in MS-se media supplemented with combinations of 2,4-D and kinetin as plant growth regulators (Table 1). Samples of somatic embryos at different developmental stages were taken from the medium with the most successful somatic embryo production (MS-se3). Torpedo-stage somatic embryos were incubated at 22–24 °C with a 16-/8-h photoperiod (irradiance of 42 µmol m–2 s–1 provided by cool-white fluorescent tubes). All media were adjusted to pH 5.8 before autoclaving (sucrose 20 g, agar 7 g). 2.2. Ultrastructural analysisFive somatic embryos of natural tetraploid T. pratense per development stage (globular, heart, torpedo, and mature) were taken from the medium and fixed with 3% glutaraldehyde (0.1 M Na-P buffer) and then 1% osmium tetroxide (0.1 M Na-P buffer) in the cold (+4 °C). The samples were dehydrated gradually by increasing concentrations of ethyl alcohol; then they were stained with uranyl acetate, saturated, and embedded into Epon 812 (Luft, 1961). For polymerization of the embedding medium, the blocks were kept in a drying oven at 30 °C for 1 night, at 45 °C for 1 day, and at 60 °C for 1 night. These blocks were then trimmed, and semithin sections were taken. Semithin sections, the first phase of TEM analyses, were stained with methylene blue and toluidine blue. The sections were examined and photographed with a DFC320 Leica camera. Ultrathin sections were stained with lead citrate (Stempak and Ward, 1964), examined with a JEOL CX 100 TEM, and micro-photographed. 2.3. Statistical analysisThe data were subjected to one-way analysis of variance (ANOVA), and the differences among means were compared by Duncan’s multiple-range test. Each treatment was replicated three times and arranged in a completely randomized design. The data given in percentages were subjected to arcsine transformation (Snedecor and Cochran, 1967) before statistical analysis.

Table 1. Experimental media used for somatic embryogenesis in natural tetraploid T. pratense.

Media 2,4-D (mg/L) Kinetin (mg/L)MS1 8 2MS2 1 2MS3 0.3 2MS4 0.5 0.5

Page 3: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

498

3. Results 3.1. Somatic embryogenesisCallus formation started within 3 days in all explants of natural tetraploid T. pratense. The calli were yellow and formed clusters in the latter days of the culture that were nodular and compact in appearance. Embryogenic calli were observed within 4–6 weeks.

No somatic embryos could be obtained from young, primary leaf explant-derived calli. Vitrification and hyperhydratation were observed in the early phase of embryogenesis in MS1. A very low rate of embryoid formation was observed in MS2 despite callus formation. The first somatic embryo formation was observed in calli in all other trials after 5–6 weeks of incubation. Globular somatic embryos were helpful for observing the differences among the media. Globular somatic embryos were formed on calli in all trials. A large number of globular somatic embryos were observed on the same callus in MS3 medium in all explants

(Figure 1a). Somatic embryos proceeding through heart and torpedo stages were monitored. Only in MS3 medium were a large number of somatic embryos able to proceed to further stages. In MS3 medium had the most successful somatic embryo production, and the average rate of hypocotyl explant-derived indirect somatic embryogenesis was 85% (Table 2). Embryos matured in 6–12 weeks. Direct somatic embryogenesis was hardly observed (Figure 1b). Somatic embryos left in the dark managed to proceed to torpedo stage but failed to proceed to mature embryo stage. When the indirect somatic embryos at torpedo stage were transferred under illumination, they were colored and matured. Samples of somatic embryos at different developmental stages were taken from MS3 medium, and ultrastructural analysis was performed (Figures 1a–1d). 3.2. Ultrastructural analysis of somatic embryos Histological and cytological analysis of indirect somatic embryos produced in vitro were conducted by

Figure 1. Globular stage (A), bar = 2 mm; heart stage (B), bar = 5 mm; torpedo stage (C), bar = 2 mm; mature stage (D), bar = 5 mm; somatic embryos of natural tetraploid T. pratense calli in MS-se3 medium.

Page 4: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

499

transmission electron microscope (TEM). Like zygotic embryos, somatic embryos proceeded through spherical (globular), heart, torpedo, and mature stages. Two types of cells were observed during various developmental stages of embryo in the semithin sections: dispersed-shaped external callus cells with translucent cytoplasm and dark-stained, uniform cells with dense cytoplasm.

In all stages of embryo development, external callus cells were nonuniform, large, sometimes oval, and generally isodiametric-shaped. Cell walls were wavy or

zig-zag shaped with translucent electrons (Figure 2a). Walls next to the intercellular spaces were thickened (Figure 2b). There was one nucleus in the center of the cell, which was small during early development and enlarged while advancing towards the mature stage. Cytoplasm formed a thin layer around the nucleus parallel to the cell walls. The nucleus was located near the center of the cell or just next to the cell wall. The nucleus had a very dark-stained and large nucleolus. The cells had large vacuoles. In some cells, vacuoles occupied almost the

Table 2. Average somatic embryogenesis (SE) rates recorded in all explants of natural tetraploid T. pratense in different media. A (apical meristem), C (cotyledon), E (epicotyl), H (hypocotyl), PL (primary leaf). Means followed by the same letter are not significantly different in same column using Duncan’s multiple range test at a 0.01 level of significance.

MediaEmbryoid/explant Average indirect SE rates (%) Direct SE

numberAbnormal observationsA C E H PL A C E H PL

MS1 1.8a 1.2a 1.6a 2.1a 1.1a 25a 18a 20a 33a 10b

MS2 2.9b 2.1a 2.4b 3.7b 2b 30a 35b 22a 65c 15b

Cotyledon-hyperhydratation,primary leaf-vitrification

MS3 4.85c 7.2b 3b 9.8c 1.2a 60b 75c 28a 85d 11b

Hypocotyl-1- Apical meristem- 1 multicotyledonary SE

Cotyledon- hyperhydratation

MS4 1.9a 1.4a 1.5a 2.2a 0.9a 31a 21a 22a 45b 0.4a

Cotyledon-somatic embryo abortion

Page 5: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

500

entire cell volume. Cytoplasm was pushed towards the cell wall. Large vacuoles located at the center of the cells contained tiny, granulated and electron-dense materials (Figure 2c). Cytoplasm pushed towards the cell wall was electron-dense in some of the cells. Cytoplasm consisted of various organelles such as starch-containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, and dictyosomes. Amyloplasts contained a large number of electron-translucent starch grains. The amyloplasts in some of the cells contained large starch grains. During the torpedo and mature stages of the somatic embryo, the number of starch grains ranged between 1 and 6 and were large. Mitochondria were elongated, oval or spherical, and electron-dense. Mitochondria were concentrated around the nucleus. In some cells, mitochondria divided by splitting in half. There were plenty of ribosomes. Dictyosomes were located adjacent to cell walls and comprised 4–6 lamellae. Granular and agranular endoplasmic reticulum (ER) was dispersed throughout the whole cytoplasm. ER was located close to the nucleus and plasma membrane. It appeared as enlarged, short structures in some areas (Figure 2d).

Embryo cells in all somatic embryo stages were dark and stained intensively. Cells were uniformly aligned. In globular and heart-stage somatic embryos, embryo cells were small and isodiametric shaped and contained a large nucleus and dense cytoplasm (Figure 3a). In torpedo and mature-stage embryos, cells were rectangular, thin-walled, and contained a large nucleus and dense cytoplasm. The walls of the cells at this stage were slightly wavy or zig-zag shaped (Figure 3b). Walls next to the intercellular spaces were thickened. In spherical and heart-stage somatic embryos, highly thickened walls occurred next to the intercellular spaces. In all somatic embryo stages, cytoplasm was dispersed homogeneously throughout the cell.

In all somatic embryo stages, the nucleus was located at the center of the cell and consisted of a spherical and dark-stained nucleolus in some of the cells. In other cells, the nucleus was pushed towards the cell wall. The nucleus had a spherical, dark-stained, and electron-dense nucleolus. Nucleoplasma was homogeneous. In some of the cells at globular somatic embryo stage, there were one or more electron-transparent areas within the nucleolus. In some

Figure 2. A. External callus cells during globular somatic embryo stage in semithin section, bar = 50 µm. B. Cell walls and intercellular spaces (indicated by arrow), bar = 3 µm. C. Electron micrograph of external callus cell, bar = 1 µm. D. Organelles in cytoplasm, bar = 5 µm (GE, globular embryo; N, nucleus; Nu, nucleolus; A, amyloplast; S, starch; r, ribosome; ER, endoplasmic reticulum; V, vacuole; M, mitochondria).

Page 6: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

501

Figure 3. A. Semithin section of somatic embryo cells, bar = 50 µm. B. Organelles in cytoplasm of embryo cells and zig-zag–shaped cell wall (indicated by arrow), bar = 1 µm. C. Nucleus membrane was wavy, vacuole-containing nucleolus in heart-stage embryo, mitochondria also occurred around the nucleus; bar = 3 µm. D. Organelles in cytoplasm, bar = 2 µm. E. Lipid bodies and zig-zag–shaped cell wall (indicated by arrow), bar = 1 µm (GE, globular embryo; N, nucleus; Nu, nucleolus; A, amyloplast; S, starch; r, ribosome; Cw, cell wall; L, lipid bodies; M, mitochondria; ER, endoplasmic reticulum; d, dictyosome).

Page 7: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

502

cells of the somatic embryos at heart-stage, vacuolization was observed at the center of the nucleolus (Figure 3c). Somatic embryos displaying such features may not be able to proceed to further stages. In heart, torpedo, and mature somatic embryo stages, the nucleus membrane was wavy or zig-zag shaped (Figure 3c). This can be seen in fast dividing cells.

During somatic embryo stages, most of the cytoplasm was located around the nucleus and was rich in organelles. There were one or more large vacuoles in the cells. Electron-dense bodies were observed in vacuoles. Large vacuoles were generally located at the center and filled with granulated material in spherical somatic embryo cells. It was remarkable that during heart, torpedo, and mature somatic embryo stages, there were storage materials in the form of dark structures along the whole tonoplast in parts of the vacuoles adjacent to the cell walls. These materials are thought to be tannin (Figure 3c).

In all somatic embryo stages, cytoplasm consisted of starch-containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, and dictyosomes. The starch grains in amyloplasts were numerous, quite large, and electron-translucent. Mitochondria were oval or spherical and electron-translucent. Mitochondrial cristae were of short length. In spherical and heart-shaped somatic embryos, mitochondria were aligned parallel to the cell wall in some cells. During torpedo and mature stages, mitochondria also occurred around the nucleus (Figures 3c and 3d).

A small number of dictyosomes were situated near the cell wall during somatic embryo stages. Granular and agranular endoplasmic reticulum (ER) was located parallel to the cell wall in all somatic embryo stages. ER was located close to the nucleus and plasma membrane. It appeared as enlarged, short structures in some areas. ER surrounded some organelles. Numerous dispersed ribosomes were observed in both free and ER-bound cytoplasm. In spherical and heart-shaped somatic embryos, there were spherical, electron-dense, large lipid bodies situated close to nucleus and parallel to the cell wall. In torpedo and mature somatic embryo stages, small lipid and protein bodies were observed. Diminishing lipid size in further stages indicates they are being used and consumed (Figure 3e).

In general, in all somatic embryo stages, external callus cells and globular and heart-stage somatic embryo cells were isodiametric, and torpedo and mature-stage embryo cells were elongated and rectangular. While external callus cells were nonuniform, somatic embryo cells were uniform in all somatic embryo stages. External callus cells and somatic embryo cells had thin and zig-zag shaped walls in all somatic embryo stages. External callus cells had dense cytoplasm situated parallel to cell walls and around the nucleus. Somatic embryo cells had electron-dense, dark-stained cytoplasm dispersed homogeneously throughout

the cell. In both types of cells, walls next to the intercellular spaces were thickened.

In most of the cells, the nucleus was located at the center of the cell and had a spherical and dark-stained nucleolus. In some of the cells, the nucleus was pushed to one side.

External callus cells had more numerous vacuoles compared to cells in all developmental stages of somatic embryos. Vacuoles contained granular, electron-dense materials. Storage materials appeared in the form of dark or gray structures along the whole tonoplast in parts of the vacuoles adjacent to the cell walls. These were probably multilamellar bodies or vesicles. One or more large vacuoles were present in somatic embryo cells. There were storage materials in the form of dark structures along the whole tonoplast in parts of the vacuoles adjacent to the cell walls. These materials are thought to be tannin. The cytoplasm contained starch-containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, and dictyosomes.

4. Discussion4.1. Somatic embryogenesisMS3 medium enabled the most successful somatic embryo production and average rate of hypocotyl explant-derived indirect somatic embryogenesis. Some embryos were degenerated at torpedo stage. Such degeneration was also recorded in zygotic embryo stages in natural tetraploid T. pratense and reported to be one of the factors causing poor seed setting (Algan and Bakar, 1996). Somatic embryos of natural tetraploid T. pratense did not exhibit a monocotyledonary or multicotyledonary structure like the somatic embryos of Glycine max and did not have very different shapes, as displayed by the somatic embryos of T. nigrescens (Fernando et al., 2002; Konieczny et al., 2012).

Plant regeneration was studied as an alternative to the seed-setting problem in natural tetraploid T. pratense; however, somatic embryogenesis has not been explored (Bakar Büyükkartal, 2008; Çölgeçen et al., 2008). A direct somatic embryogenesis rate of approximately 40% was achieved in the studies in diploid (2n = 14) T. pratense varieties by using a variety of culture methods and nutrient media containing 2,4-D, BAP, kinetin, yeast, or casein hydrolysate. In the present study, an 85% indirect somatic embryogenesis rate was achieved with 2,4-D + kinetin combinations. Seedling pieces, zygotic embryos, and mesophyll protoplasts were also used in other studies. In this study, for the first time high rates of somatic embryogenesis were recorded in all explants of 15-day-old aseptic seedlings, except in the young primary leaf of T. pratense L. It was confirmed that young aseptic seedlings need to be used for somatic embryogenesis in natural tetraploid T. pratense, as reported in other studies

Page 8: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

503

(Phillips and Collins, 1980; Maheswaran and Williams, 1984, 1985, 1986; Radionenko et al., 1994). In the studies on diploid T. pratense, the most successful results were obtained from SL medium supplemented with 0.01 mg/L 2,4-D + 2 mg/L adenin, EC6 medium supplemented with 0.05 mg/L BAP, and KM8p medium supplemented with 0.5 mg/L 2,4-D + 0.5 mg/L kinetin. In the present study, MS3 medium supplemented with 0.3 mg/L 2,4-D + 2 mg/L kinetin proved to be the most successful medium (Phillips and Collins, 1980; Maheswaran and Williams, 1984, 1985, 1986; Radionenko et al., 1994; Mclean and Nowak, 1998). Stimulation of somatic embryogenesis by high concentrations of 2,4-D has not been confirmed in every plant species (Babaoğlu et al., 2001). In a study on T. nigrescens, 2,4-D was more effective for an inductive effect than NAA. However, in this study, the 2,4-D + kinetin combination proved to be successful. Combined use of auxin and cytokinin supports normal somatic embryo development (Konieczny et al., 2010). In the present study, the combined low concentrations of 2,4-D with 2 mg/L kinetin stimulated somatic embryogenesis. Proportionally, kinetin was effective at higher concentrations compared to 2,4-D. An KM8p protoplast culture medium with a 0.5 mg/L 2,4-D + 0.5 mg/L kinetin (Radionenko et al., 1994) combination resulted in low somatic embryogenesis rates in this study. Yeast extract and casein hydrolysate were used in other research, but they were not used in the trials of our study. Culture medium was also used for subculture, and somatic embryos were matured. No embryonic structural abnormalities (multicotyledonary somatic embryos), as observed in T. nigrenses, were recorded; however, some of the embryos could not develop into mature somatic embryos (Phillips and Collins, 1980; Maheswaran and Williams, 1984, 1985, 1986; Radionenko et al., 1994; Konieczny et al., 2012). There are studies available in the literature on the ultrastructural analysis of somatic embryogenesis of Trifolium and other species. Proembryos or somatic embryos were formed in callus cultures from hypocotyl sections of Trifolium incarnatum L. and T. ambiguum Bieb. on Phillips and Collins L2 medium with 0.06 mg/L picloram and 0.1 mg/L 6-benzylaminopurine and T. repens L. and T. vesiculosum Savi. on Gamborg B5 medium with 1.25 mg/L 2,4-dichlorophenoxyacetic acid (2.4-D), 0.5 mg/L naphthaleneacetic acid (NAA), and 0.5 mg/L 6-furfurylaminopurine (kinetin) (Pederson, 1986). Media and plant growth regulators other than those used in our trials were used in these studies. In a departure from our study, immature zygotic embryos at torpedo and cotyledonary stage were cultured on MS media with 2,4-D in T. nigrescens (Viv.) for somatic embryogenesis (Konieczny et al., 2010). In addition, somatic embryogenesis was realized with immature

zygotic embryos, leaf protoplasts of Arabidopsis, and transformed cotton by Agrobacterium (Luo et al., 1997; Shang et al., 2009).4.2. Ultrastructural analysis of somatic embryos Two types of cells were observed in somatic embryogenesis in the semithin sections: dispersed-shaped external callus cells and uniform cells. A single-row epidermal layer and 5–6 rows of cortical cells were observed during embryo development in direct somatic embryogenesis of T. repens. Cortical cells were smaller and meristematic. The cells had dense cytoplasm, tiny starch grains and vacuoles, and a visible nucleus and nucleolus. A single row of radially oriented epidermis cells with cortical cells beneath were observed in somatic embryos of Quercus suber, but they did not show any differences. The cells beneath external callus cells were similar in appearance in the present study. Cytoplasm consisted of various organelles such as starch-containing amyloplasts, mitochondria, plastids, ribosomes, endoplasmic reticulum, and dictyosomes. Subcellular structures did not exhibit any differences (Maheswaran and Willams, 1985; Puigderrajols et al., 2001).

Similar to the findings of this study, the presence of thin cell walls, a large nucleus, dense cytoplasm, large vacuoles, abundant starch, and other organelles was reported in former ultrastructural studies of other species in somatic embryos (Karlson and Vasil, 1986; Brisribe et al., 1993; Jasik et al., 1995; Puigderrajols et al., 2001). Vacuolization was also observed in the nucleolus of wheat somatic embryos (Wang et al., 1994). No lipids or proteins were reported in Acrocomia aculeata somatic embryos despite the presence of large vacuoles with electron-dense materials. This was considered an indicator of failure in embryo maturation by researchers (Moura et al., 2010). In the present study, despite their decreasing amounts in torpedo and mature stages, lipids and proteins were detectable in somatic embryos, as reported for Q. suber. Somatic embryos were matured. The decrease in storage materials in further developmental stages was attributed to increased mitotic activity and the differentiation and maturation process of the embryo (Puigderrajols et al., 2001). There are no studies on the ultrastructure of diploid or tetraploid Trifolium pratense somatic embryos.

First, somatic embryos of natural tetraploid T. pratense appeared within 5–6 weeks on the calli obtained from apical meristem, hypocotyl, epicotyl, and cotyledon explants planted on petri dishes. Explants produced embryo-generating calli. Globular, heart, torpedo, and mature-stage somatic embryos were mostly produced in MS3 medium, and their ultrastructure was examined. The combination of 0.3 mg/L 2,4-D and 2 mg/L kinetin was favorable for somatic embryogenesis of natural tetraploid T. pratense from all explants except young primary leaf. Hence, somatic embryogenesis can be considered as a new

Page 9: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

504

plant regeneration method and an alternative to the seed-setting problem in this valuable plant. Our study developed an effective and efficient in vitro production method for using natural tetraploid T. pratense in biotechnological studies. The somatic embryos of natural tetraploid T. pratense were produced from hypocotyl (85%), cotyledon (75%), and apical meristem (60%) explants. Non-uniform external callus cells with translucent cytoplasm and uniformly ordered, dark-stained embryo cells with dense cytoplasm were observed in developmental stages of somatic embryos. The presence of vacuoles or transparent zones in the nucleoli during globular and heart stages indicates a failure of some somatic embryos to complete

maturation. Despite similar developmental stages and cell characteristics of zygotic and somatic embryos, the walls of somatic embryo cells revealed a highly wavy pattern. The present study is expected to promote future studies on the implementation of advanced biotechnological techniques and synthetic seed production.

Acknowledgments This study was supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK; project number: 108T830; TBAG). The authors would like to extend their gratitude to Prof Dr Şahabettin Elçi for kindly providing the T. pratense seeds for this study.

References

Algan G, Bakar HN (1996). Light and electron microscopic examination of the embryo and endosperm development in the natural tetraploid Trifolium pratense L. Isr J Plant Sci 44: 273-288.

Algan G, Bakar HN (1997). The ultrastructure of the mature embryo sac in the natural tetraploid of red clover (Trifolium pratense L.) that has a very low rate of seed formation. Acta Soc Bot Pol 66: 13-20.

Algan G, Bakar Büyükkartal HN (1999a). Ultrastructure of seed coat development in the natural tetraploid Trifolium pratense L. J Agron Crop Sci 184: 205-213.

Algan G, Bakar Büyükkartal HN (1999b). Tetraploid çayırüçgülü (Trifolium pratense L.) ’nde apomiktik gelişme. Turk J Agric For 23: 519-525 (article in Turkish with an abstract in English).

Babaoğlu M, Gürel E, Ozcan S (2001). Bitki Biyoteknolojisi I. Doku Kültürü ve Uygulamaları. Birinci Baskı, Konya, Türkiye: Selcuk Universitesi Basımevi (in Turkish).

Bakar Büyükkartal HN (2008). Causes of low seed set in the natural tetraploid Trifolium pratense L. (Fabaceae). Afr J Biotechnol 7: 1240-1249.

Bakar HN, Algan G (1998). Ultrastructure of the endosperm haustorium in Trifolium pratense L. Acta Biol Cracov Bot 40: 15-19.

Brisibe EA, Nishioka D, Miyake H, Taniguchi T, Maeda E (1993). Developmental electron microscopy and histochemistry of somatic embryo differentiation in sugarcane. Plant Sci 89(1): 85-92.

Çölgeçen H, Büyükkartal HN, Toker MC (2008). In vitro germination and structure of hard seed testa of natural tetraploid Trifolium pratense L. Afr J Biotechnol 7: 1473-1478.

Çölgeçen H, Toker MC (2008). Plant regeneration of natural tetraploid Trifolium pratense L. Biol Res 41: 25-31.

Elçi Ş (1982). The utilization of genetic resource in fodder crop breeding. Eucarpio, Fodder Crop Section, 13–16 September, Aberystwyth, UK, pp. 28.

Fernando JA, Vieria MLC, Geraldi IO, Gloria BA (2002). Anatomical study of somatic embryogenesis in Glycine max (L.) Merrill. Braz Arch Biol Techn 45(3): 277-286.

Jasik J, Salajova T, Salaj J (1995). Developmental anatomy and ultrastructure of early somatic embryos in European black pine (Pinus nigra Arn.). Protoplasma 185: 205-211.

Karlson SB, Basil IK (1986). Morphology and ultrastructure of embryogenic cell suspension cultures of Panicum maximum (guinea grass) and Pennisetum purpureum (napier grass). Am J Bot 73(6): 894-901.

Konieczny R, Pilarska M, Tuleja M, Salaj T, Ilnicki T (2010). Somatic embryogenesis and plant regeneration in zygotic embryos of Trifolium nigrescens (Viv.). Plant Cell Tiss Org 100: 123-130.

Konieczny R, Sliwinska E, Pilarska M, Tuleja M (2012). Morphohistological and flow cytometric analyses of somatic embryogenesis in Trifolium nigrescens Viv. Plant Cell Tiss Org 109: 131-141.

Luft JH (1961). Improvements in epoxy resin embedding methods. J Biophys Biochem Cy 9: 409.

Luo Y, Koop HU (1997). Somatic embryogenesis in cultured immature zygotic embryos and leaf protoplasts of Arabidopsis thaliana ecotypes. Planta 202: 387-396.

Maheswaran G, William EG (1984). Direct somatic embryoid formation on immature embryos of Trifolium repens, T. pratense and Medicago sativa and rapid clonal propagation of T. repens. Ann Bot-London 54: 201-211.

Maheswaran G, William EG (1985). Origin and development of somatic embryoids formed directly on immature embryos of Trifolium repens in vitro. Ann Bot-London 56: 619-630.

Maheswaran G, William EG (1986). Clonal propagation of T. pratense, T. resupinatum and T. subterraneum by direct somatic embryogenesis on cultured immature embryos. Plant Cell Rep 3: 165-168.

Moura EF, Ventrella MC, Motoike SY (2010). Anatomy, histochemistry and ultrastructure of seed and somatic embryo of Acrocomia aculeata (Arecaceae). Sci Agr (Piracicaba, Braz.) 67: 399-407.

Page 10: Ultrastructural observations in somatic embryogenesis of ...journals.tubitak.gov.tr/botany/issues/bot-16-40-5/... · plant regeneration is somatic embryogenesis. Although, there is

ÇÖLGEÇEN et al. / Turk J Bot

505

Mclean NL, Nowak J (1998). Inheritance of somatic embryogenesis in red clover (Trifolium pratense L.). Theor Appl Genet 97: 557-562.

Monteiro M, Appezzato-da-Gloria B, Valarini MJ, Alberto de Oliviera C, Vieira MLC (2003). Plant regeneration from protoplast of alfalfa (Medicago sativa) via somatic embryogenesis. Sci Agr 60: 683-689.

Murashige T, Skoog FA (1962). Revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiol Plantarum 15: 473-497.

Pederson GA (1986). In vitro culture and somatic embryogenesis of four Trifolium species. Plant Sci 45: 101-104.

Phillips GC, Collins GB (1980). Somatic embryogenesis from cell suspension cultures of red clover. Crop Sci 20: 323-326.

Puigderrajols P, Mir G, Molinas M (2001). Ultrastructure of early secondary embryogenesis by multicellular and unicellular pathways in cork oak (Quercus suber L.). Ann Bot-London 87: 000-011.

Radionenko MA, Kuchuk NV, Khvedynich OA, Gleba YY (1994). Direct somatic embryogenesis and plant regeneration from protoplast of red clover (Trifolium pratense L.). Plant Sci 97: 75-81.

Ramakrishnan K, Granam R, Sivakumar P, Manickani A (2005). Developmental pattern formation of somatic embryos induced in cell suspension cultures of cowpea (Vigna unguiculata (L.) Walp). Plant Cell Rep 24: 501-506.

Shang H, Liu C, Zhang C, Li F, Hong W, Li FG (2009). Histological and ultrastructural observation reveals significant cellular differences between Agrobacterium transformed embryogenic and non-embryogenic calli of cotton. J Integr Plant Biol 7: 1-10.

Snedecor GW, Cochran WG (1967). Statistical Methods. 6th ed. Iowa, USA: The Iowa State University Press, pp. 327-329.

Stempak JG, Ward RT (1964). An improved staining method for electron microscopy. J Cell Biol 22: 697.

Wang YF, Cui KR, Wang LH, Wang LS (1994). Ultrastructural studies of somatic embryogenesis in wheat. Acta Bot Sin 36: 418-422.